UMass-Foundation-Model / 3D-VLA

[ICML 2024] 3D-VLA: A 3D Vision-Language-Action Generative World Model
https://vis-www.cs.umass.edu/3dvla/
320 stars 11 forks source link

3D Embodied Instruction Tuning Dataset #6

Open zubair-irshad opened 2 months ago

zubair-irshad commented 2 months ago

Hi Authors,

Great work, and thanks for releasing your code. Do you plan on releasing your 3D Embodied Instruction Tuning Dataset or code for us to process the dataset from existing robotics and HOI dataset mentioned in your paper? Any help is appreciated, thanks!